sparsegl: Sparse Group Lasso

Efficient implementation of sparse group lasso with optional bound constraints on the coefficients; see <doi:10.18637/jss.v110.i06>. It supports the use of a sparse design matrix as well as returning coefficient estimates in a sparse matrix. Furthermore, it correctly calculates the degrees of freedom to allow for information criteria rather than cross-validation with very large data. Finally, the interface to compiled code avoids unnecessary copies and allows for the use of long integers.

Version: 1.1.1
Depends: R (≥ 3.5)
Imports: cli, dotCall64, ggplot2, magrittr, Matrix, methods, rlang, RSpectra, tidyr
Suggests: dplyr, gglasso, glmnet, knitr, markdown, rmarkdown, splines, testthat (≥ 3.0.0)
Published: 2024-08-24
DOI: 10.32614/CRAN.package.sparsegl
Author: Daniel J. McDonald [aut, cre], Xiaoxuan Liang [aut], Anibal Solón Heinsfeld [aut], Aaron Cohen [aut], Yi Yang [ctb], Hui Zou [ctb], Jerome Friedman [ctb], Trevor Hastie [ctb], Rob Tibshirani [ctb], Balasubramanian Narasimhan [ctb], Kenneth Tay [ctb], Noah Simon [ctb], Junyang Qian [ctb], James Yang [ctb]
Maintainer: Daniel J. McDonald <daniel at stat.ubc.ca>
BugReports: https://github.com/dajmcdon/sparsegl/issues
License: MIT + file LICENSE
URL: https://github.com/dajmcdon/sparsegl, https://dajmcdon.github.io/sparsegl/
NeedsCompilation: yes
Citation: sparsegl citation info
Materials: README NEWS
CRAN checks: sparsegl results

Documentation:

Reference manual: sparsegl.pdf
Vignettes: Getting started with sparsegl (source, R code)

Downloads:

Package source: sparsegl_1.1.1.tar.gz
Windows binaries: r-devel: sparsegl_1.1.1.zip, r-release: sparsegl_1.1.1.zip, r-oldrel: sparsegl_1.1.1.zip
macOS binaries: r-release (arm64): sparsegl_1.1.1.tgz, r-oldrel (arm64): sparsegl_1.1.1.tgz, r-release (x86_64): sparsegl_1.1.1.tgz, r-oldrel (x86_64): sparsegl_1.1.1.tgz
Old sources: sparsegl archive

Reverse dependencies:

Reverse depends: absorber

Linking:

Please use the canonical form https://CRAN.R-project.org/package=sparsegl to link to this page.